File size: 63,316 Bytes
edf1149 526a1d2 edf1149 69256da bd3410e 520de88 526a1d2 520de88 526a1d2 520de88 edf1149 526a1d2 edf1149 520de88 edf1149 520de88 edf1149 520de88 edf1149 520de88 edf1149 69256da edf1149 69256da 520de88 69256da 526a1d2 edf1149 520de88 edf1149 526a1d2 edf1149 526a1d2 520de88 edf1149 526a1d2 edf1149 69256da 526a1d2 520de88 edf1149 520de88 edf1149 520de88 edf1149 526a1d2 edf1149 69256da edf1149 69256da 526a1d2 edf1149 69256da 526a1d2 edf1149 3b1e6c7 edf1149 520de88 edf1149 69256da edf1149 526a1d2 520de88 edf1149 526a1d2 edf1149 526a1d2 edf1149 520de88 526a1d2 edf1149 520de88 edf1149 520de88 526a1d2 edf1149 526a1d2 edf1149 526a1d2 edf1149 520de88 526a1d2 520de88 526a1d2 edf1149 520de88 526a1d2 520de88 526a1d2 520de88 526a1d2 520de88 526a1d2 520de88 526a1d2 520de88 526a1d2 520de88 526a1d2 520de88 526a1d2 520de88 526a1d2 520de88 526a1d2 520de88 526a1d2 520de88 526a1d2 edf1149 520de88 526a1d2 520de88 526a1d2 520de88 526a1d2 bd3410e 526a1d2 520de88 edf1149 526a1d2 520de88 edf1149 526a1d2 edf1149 526a1d2 520de88 526a1d2 520de88 526a1d2 bd3410e 526a1d2 edf1149 526a1d2 edf1149 526a1d2 520de88 edf1149 69256da 526a1d2 520de88 526a1d2 bd3410e 526a1d2 edf1149 526a1d2 520de88 526a1d2 520de88 526a1d2 edf1149 526a1d2 520de88 526a1d2 bd3410e 526a1d2 520de88 edf1149 526a1d2 520de88 526a1d2 520de88 526a1d2 bd3410e 526a1d2 edf1149 520de88 edf1149 526a1d2 edf1149 520de88 526a1d2 520de88 526a1d2 edf1149 526a1d2 edf1149 526a1d2 edf1149 526a1d2 edf1149 526a1d2 520de88 526a1d2 edf1149 526a1d2 edf1149 526a1d2 edf1149 526a1d2 bd3410e 526a1d2 edf1149 520de88 526a1d2 520de88 526a1d2 edf1149 520de88 526a1d2 520de88 526a1d2 520de88 526a1d2 520de88 526a1d2 520de88 526a1d2 edf1149 520de88 526a1d2 edf1149 526a1d2 520de88 526a1d2 520de88 526a1d2 edf1149 526a1d2 edf1149 526a1d2 edf1149 526a1d2 edf1149 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 |
# DEPENDENCIES
import re
import json
from typing import Any
from typing import Dict
from typing import List
from pathlib import Path
from loguru import logger
from typing import Optional
from datetime import datetime
from dataclasses import dataclass
from detector.orchestrator import DetectionResult
from detector.attribution import AttributionResult
from reporter.reasoning_generator import DetailedReasoning
from reporter.reasoning_generator import ReasoningGenerator
@dataclass
class DetailedMetric:
"""
Metric data structure with sub-metrics
"""
name : str
ai_probability : float
human_probability : float
confidence : float
verdict : str
description : str
detailed_metrics : Dict[str, float]
weight : float
class ReportGenerator:
"""
Generates comprehensive detection reports with detailed metrics
Supports:
- JSON (structured data with all details)
- PDF (printable reports with tables and formatting)
"""
def __init__(self, output_dir: Optional[Path] = None):
"""
Initialize report generator
Arguments:
----------
output_dir { str } : Directory for saving reports (default: data/reports)
"""
if (output_dir is None):
output_dir = Path(__file__).parent.parent / "data" / "reports"
self.output_dir = Path(output_dir)
self.output_dir.mkdir(parents = True,
exist_ok = True,
)
self.reasoning_generator = ReasoningGenerator()
logger.info(f"ReportGenerator initialized (output_dir={self.output_dir})")
def generate_complete_report(self, detection_result: DetectionResult, attribution_result: Optional[AttributionResult] = None, highlighted_sentences: Optional[List] = None,
formats: List[str] = ["json", "pdf"], filename_prefix: str = "ai_detection_report") -> Dict[str, str]:
"""
Generate comprehensive report in JSON and PDF formats with detailed metrics
Arguments:
----------
detection_result : Detection analysis result
attribution_result : Model attribution result (optional)
highlighted_sentences : List of highlighted sentences (optional)
formats : List of formats to generate (json, pdf)
filename_prefix : Prefix for output filenames
Returns:
--------
{ dict } : Dictionary mapping format to filepath
"""
# Convert DetectionResult to dict for consistent access
detection_dict = detection_result.to_dict() if hasattr(detection_result, 'to_dict') else detection_result
# Extract the actual detection data from the structure
if ("detection_result" in detection_dict):
detection_data = detection_dict["detection_result"]
logger.info("Extracted detection_result from outer dict")
else:
detection_data = detection_dict
logger.info("Using detection_dict directly")
# Generate detailed reasoning
reasoning = self.reasoning_generator.generate(ensemble_result = detection_result.ensemble_result,
metric_results = detection_result.metric_results,
domain = detection_result.domain_prediction.primary_domain,
attribution_result = attribution_result,
text_length = detection_result.processed_text.word_count,
)
# Extract detailed metrics from ACTUAL detection results
detailed_metrics = self._extract_detailed_metrics(detection_data = detection_data)
# Timestamp for filenames
timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")
generated_files = dict()
# Generate requested formats
if ("json" in formats):
json_path = self._generate_json_report(detection_data = detection_data,
detection_dict_full = detection_dict,
reasoning = reasoning,
detailed_metrics = detailed_metrics,
attribution_result = attribution_result,
highlighted_sentences = highlighted_sentences,
filename = f"{filename_prefix}_{timestamp}.json",
)
generated_files["json"] = str(json_path)
if ("pdf" in formats):
try:
pdf_path = self._generate_pdf_report(detection_data = detection_data,
detection_dict_full = detection_dict,
reasoning = reasoning,
detailed_metrics = detailed_metrics,
attribution_result = attribution_result,
highlighted_sentences = highlighted_sentences,
filename = f"{filename_prefix}_{timestamp}.pdf",
)
generated_files["pdf"] = str(pdf_path)
except Exception as e:
logger.warning(f"PDF generation failed: {repr(e)}")
logger.info("Install reportlab for PDF support: pip install reportlab")
logger.info(f"Generated {len(generated_files)} report(s): {list(generated_files.keys())}")
return generated_files
def _extract_detailed_metrics(self, detection_data: Dict) -> List[DetailedMetric]:
"""
Extract detailed metrics with sub-metrics from ACTUAL detection result
"""
detailed_metrics = list()
metrics_data = detection_data.get("metrics", {})
ensemble_data = detection_data.get("ensemble", {})
# Get actual metric weights from ensemble
metric_weights = ensemble_data.get("metric_contributions", {})
# Extract actual metric data
for metric_name, metric_result in metrics_data.items():
if (not isinstance(metric_result, dict)):
logger.warning(f"Metric {metric_name} is not a dict: {type(metric_result)}")
continue
if (metric_result.get("error") is not None):
logger.warning(f"Metric {metric_name} has error: {metric_result.get('error')}")
continue
# Get actual probabilities and confidence
ai_prob = metric_result.get("ai_probability", 0)
human_prob = metric_result.get("human_probability", 0)
confidence = metric_result.get("confidence", 0)
# Determine verdict based on actual probability
if (human_prob >= 0.6):
verdict = "HUMAN"
elif (ai_prob >= 0.6):
verdict = "AI"
elif (ai_prob > 0.4 and ai_prob < 0.6):
verdict = "MIXED"
elif (human_prob > 0.4 and human_prob < 0.6):
verdict = "MIXED"
else:
# If both low, check which is higher
if (human_prob > ai_prob):
verdict = "HUMAN"
elif (ai_prob > human_prob):
verdict = "AI"
else:
verdict = "MIXED"
# Get actual weight or use default
weight = 0.0
if (metric_name in metric_weights):
weight = metric_weights[metric_name].get("weight", 0.0)
# Extract actual detailed metrics from metric result
detailed_metrics_data = self._extract_metric_details(metric_name = metric_name,
metric_result = metric_result,
)
# Get description based on metric type
description = self._get_metric_description(metric_name = metric_name)
detailed_metrics.append(DetailedMetric(name = metric_name,
ai_probability = ai_prob * 100, # Convert to percentage
human_probability = human_prob * 100, # Convert to percentage
confidence = confidence * 100, # Convert to percentage
verdict = verdict,
description = description,
detailed_metrics = detailed_metrics_data,
weight = weight * 100, # Convert to percentage
)
)
logger.info(f"Extracted {len(detailed_metrics)} detailed metrics")
return detailed_metrics
def _extract_metric_details(self, metric_name: str, metric_result: Dict) -> Dict[str, float]:
"""
Extract detailed sub-metrics from metric result
"""
details = dict()
# Try to get details from metric result
if metric_result.get("details"):
# Extract all numeric details
for key, value in metric_result["details"].items():
if (isinstance(value, (int, float))):
# Format specific metrics appropriately
if ("perplexity" in key.lower()):
details[key] = float(f"{value:.2f}")
elif ("entropy" in key.lower()):
details[key] = float(f"{value:.2f}")
elif (("score" in key.lower()) or ("ratio" in key.lower())):
details[key] = float(f"{value:.4f}")
elif ("probability" in key.lower()):
details[key] = float(f"{value:.4f}")
else:
details[key] = float(f"{value:.3f}")
else:
details[key] = value
# If no details available, provide basic calculated values
if not details:
details = {"ai_probability" : metric_result.get("ai_probability", 0) * 100,
"human_probability" : metric_result.get("human_probability", 0) * 100,
"confidence" : metric_result.get("confidence", 0) * 100,
"score" : metric_result.get("raw_score", 0) * 100,
}
return details
def _get_metric_description(self, metric_name: str) -> str:
"""
Get description for each metric type
"""
descriptions = {"structural" : "Analyzes sentence structure, length patterns, and statistical features",
"perplexity" : "Measures text predictability using language model cross-entropy",
"entropy" : "Evaluates token diversity and sequence unpredictability",
"semantic_analysis" : "Examines semantic coherence, topic consistency, and logical flow",
"linguistic" : "Assesses grammatical patterns, syntactic complexity, and style markers",
"multi_perturbation_stability" : "Tests text stability under perturbation using curvature analysis",
}
return descriptions.get(metric_name, "Advanced text analysis metric.")
def _generate_json_report(self, detection_data: Dict, detection_dict_full: Dict, reasoning: DetailedReasoning, detailed_metrics: List[DetailedMetric],
attribution_result: Optional[AttributionResult], highlighted_sentences: Optional[List] = None, filename: str = None) -> Path:
"""
Generate JSON format report with detailed metrics
"""
# Convert metrics to serializable format
metrics_data = list()
for metric in detailed_metrics:
metrics_data.append({"name" : metric.name,
"ai_probability" : metric.ai_probability,
"human_probability" : metric.human_probability,
"confidence" : metric.confidence,
"verdict" : metric.verdict,
"description" : metric.description,
"weight" : metric.weight,
"detailed_metrics" : metric.detailed_metrics,
})
# Convert highlighted sentences to serializable format
highlighted_data = None
if highlighted_sentences:
highlighted_data = list()
for sent in highlighted_sentences:
highlighted_data.append({"text" : sent.text,
"ai_probability" : sent.ai_probability,
"confidence" : sent.confidence,
"color_class" : sent.color_class,
"index" : sent.index,
})
# Attribution data
attribution_data = None
if attribution_result:
attribution_data = {"predicted_model" : attribution_result.predicted_model.value,
"confidence" : attribution_result.confidence,
"model_probabilities" : attribution_result.model_probabilities,
"reasoning" : attribution_result.reasoning,
"fingerprint_matches" : attribution_result.fingerprint_matches,
"domain_used" : attribution_result.domain_used.value,
"metric_contributions": attribution_result.metric_contributions,
}
# Use detection results from dictionary
ensemble_data = detection_data.get("ensemble", {})
analysis_data = detection_data.get("analysis", {})
metrics_data_dict = detection_data.get("metrics", {})
performance_data = detection_data.get("performance", {})
report_data = {"report_metadata" : {"generated_at" : datetime.now().isoformat(),
"version" : "1.0.0",
"format" : "json",
"report_id" : filename.replace('.json', ''),
},
"overall_results" : {"final_verdict" : ensemble_data.get("final_verdict", "Unknown"),
"ai_probability" : ensemble_data.get("ai_probability", 0),
"human_probability" : ensemble_data.get("human_probability", 0),
"mixed_probability" : ensemble_data.get("mixed_probability", 0),
"overall_confidence" : ensemble_data.get("overall_confidence", 0),
"uncertainty_score" : ensemble_data.get("uncertainty_score", 0),
"consensus_level" : ensemble_data.get("consensus_level", 0),
"domain" : analysis_data.get("domain", "general"),
"domain_confidence" : analysis_data.get("domain_confidence", 0),
"text_length" : analysis_data.get("text_length", 0),
"sentence_count" : analysis_data.get("sentence_count", 0),
},
"ensemble_analysis" : {"method_used" : "confidence_calibrated",
"metric_weights" : ensemble_data.get("metric_contributions", {}),
"reasoning" : ensemble_data.get("reasoning", []),
},
"detailed_metrics" : metrics_data,
"detection_reasoning" : {"summary" : reasoning.summary,
"key_indicators" : reasoning.key_indicators,
"metric_explanations" : reasoning.metric_explanations,
"supporting_evidence" : reasoning.supporting_evidence,
"contradicting_evidence" : reasoning.contradicting_evidence,
"confidence_explanation" : reasoning.confidence_explanation,
"domain_analysis" : reasoning.domain_analysis,
"ensemble_analysis" : reasoning.ensemble_analysis,
"uncertainty_analysis" : reasoning.uncertainty_analysis,
"recommendations" : reasoning.recommendations,
},
"highlighted_text" : highlighted_data,
"model_attribution" : attribution_data,
"performance_metrics" : {"total_processing_time" : performance_data.get("total_time", 0),
"metrics_execution_time" : performance_data.get("metrics_time", {}),
"warnings" : detection_data.get("warnings", []),
"errors" : detection_data.get("errors", []),
}
}
output_path = self.output_dir / filename
with open(output_path, 'w', encoding='utf-8') as f:
json.dump(obj = report_data,
fp = f,
indent = 4,
ensure_ascii = False,
)
logger.info(f"JSON report saved: {output_path}")
return output_path
def _generate_pdf_report(self, detection_data: Dict, detection_dict_full: Dict, reasoning: DetailedReasoning, detailed_metrics: List[DetailedMetric],
attribution_result: Optional[AttributionResult], highlighted_sentences: Optional[List] = None, filename: str = None) -> Path:
"""
Generate PDF format report with detailed metrics
"""
try:
from reportlab.lib import colors
from reportlab.lib.units import cm
from reportlab.platypus import Table
from reportlab.lib.units import inch
from reportlab.platypus import Spacer
from reportlab.lib.pagesizes import A4
from reportlab.lib.enums import TA_LEFT
from reportlab.platypus import PageBreak
from reportlab.platypus import Paragraph
from reportlab.lib.enums import TA_RIGHT
from reportlab.graphics import renderPDF
from reportlab.lib.enums import TA_CENTER
from reportlab.platypus import TableStyle
from reportlab.pdfgen.canvas import Canvas
from reportlab.lib.enums import TA_JUSTIFY
from reportlab.lib.pagesizes import letter
from reportlab.graphics.shapes import Line
from reportlab.graphics.shapes import Rect
from reportlab.platypus import KeepTogether
from reportlab.graphics.shapes import Circle
from reportlab.graphics.shapes import Drawing
from reportlab.lib.styles import ParagraphStyle
from reportlab.platypus import SimpleDocTemplate
from reportlab.graphics.charts.piecharts import Pie
from reportlab.platypus.flowables import HRFlowable
from reportlab.lib.styles import getSampleStyleSheet
from reportlab.graphics.charts.textlabels import Label
from reportlab.graphics.widgets.markers import makeMarker
except ImportError:
raise ImportError("reportlab is required for PDF generation. Install: pip install reportlab")
output_path = self.output_dir / filename
# Create PDF with pre-defined settings
doc = SimpleDocTemplate(str(output_path),
pagesize = A4,
rightMargin = 0.75*inch,
leftMargin = 0.75*inch,
topMargin = 0.75*inch,
bottomMargin = 0.75*inch,
)
# Container for PDF elements
elements = list()
styles = getSampleStyleSheet()
# Color Scheme
PRIMARY_COLOR = '#3b82f6' # Blue-600
SUCCESS_COLOR = '#10b981' # Emerald-500
WARNING_COLOR = '#f59e0b' # Amber-500
DANGER_COLOR = '#ef4444' # Red-500
INFO_COLOR = '#8b5cf6' # Violet-500
GRAY_LIGHT = '#f8fafc' # Gray-50
GRAY_MEDIUM = '#e2e8f0' # Gray-200
GRAY_DARK = '#334155' # Gray-700
TEXT_COLOR = '#1e293b' # Gray-800
# Custom Styles
title_style = ParagraphStyle('PremiumTitle',
parent = styles['Heading1'],
fontName = 'Helvetica-Bold',
fontSize = 28,
textColor = PRIMARY_COLOR,
spaceAfter = 20,
alignment = TA_CENTER,
)
subtitle_style = ParagraphStyle('PremiumSubtitle',
parent = styles['Normal'],
fontName = 'Helvetica',
fontSize = 12,
textColor = GRAY_DARK,
spaceAfter = 30,
alignment = TA_CENTER,
)
filename_style = ParagraphStyle('FilenameStyle',
parent = styles['Normal'],
fontName = 'Helvetica-Bold',
fontSize = 10,
textColor = GRAY_DARK,
spaceAfter = 10,
alignment = TA_CENTER,
)
section_style = ParagraphStyle('PremiumSection',
parent = styles['Heading2'],
fontName = 'Helvetica-Bold',
fontSize = 18,
textColor = TEXT_COLOR,
spaceAfter = 12,
spaceBefore = 20,
underlineWidth = 1,
underlineColor = PRIMARY_COLOR,
)
subsection_style = ParagraphStyle('PremiumSubSection',
parent = styles['Heading3'],
fontName = 'Helvetica-Bold',
fontSize = 14,
textColor = GRAY_DARK,
spaceAfter = 8,
spaceBefore = 16,
)
key_indicators_style = ParagraphStyle('KeyIndicatorsStyle',
parent = styles['Heading2'],
fontName = 'Helvetica-Bold',
fontSize = 18,
textColor = TEXT_COLOR,
spaceAfter = 12,
spaceBefore = 20,
underlineWidth = 1,
underlineColor = PRIMARY_COLOR,
)
body_style = ParagraphStyle('PremiumBody',
parent = styles['BodyText'],
fontName = 'Helvetica',
fontSize = 11,
textColor = TEXT_COLOR,
alignment = TA_JUSTIFY,
spaceAfter = 8,
)
# Larger font for page 2 content
page2_body_style = ParagraphStyle('Page2Body',
parent = styles['BodyText'],
fontName = 'Helvetica',
fontSize = 11,
textColor = TEXT_COLOR,
alignment = TA_JUSTIFY,
spaceAfter = 8,
)
bullet_style = ParagraphStyle('BulletStyle',
parent = styles['BodyText'],
fontName = 'Helvetica',
fontSize = 11,
textColor = TEXT_COLOR,
alignment = TA_LEFT,
spaceAfter = 6,
leftIndent = 20,
)
bold_style = ParagraphStyle('BoldStyle',
parent = styles['BodyText'],
fontName = 'Helvetica-Bold',
fontSize = 11,
textColor = TEXT_COLOR,
alignment = TA_LEFT,
spaceAfter = 8,
)
small_bold_style = ParagraphStyle('SmallBoldStyle',
parent = styles['BodyText'],
fontName = 'Helvetica-Bold',
fontSize = 9,
textColor = TEXT_COLOR,
alignment = TA_LEFT,
spaceAfter = 4,
)
small_style = ParagraphStyle('SmallStyle',
parent = styles['BodyText'],
fontName = 'Helvetica',
fontSize = 9,
textColor = TEXT_COLOR,
alignment = TA_LEFT,
spaceAfter = 4,
)
footer_style = ParagraphStyle('FooterStyle',
parent = styles['Normal'],
fontName = 'Helvetica',
fontSize = 9,
textColor = GRAY_DARK,
alignment = TA_CENTER,
)
print (detection_dict_full.keys())
# Use detection results from detection_data
ensemble_data = detection_data.get("ensemble", {})
analysis_data = detection_data.get("analysis", {})
performance_data = detection_data.get("performance", {})
# Extract filename from file_info
file_info = detection_data.get("file_info", {})
# Extract Analyzed File name from file_info
original_filename = file_info.get("filename", "Unknown")
# Extract values - handle different data formats
ai_prob = ensemble_data.get("ai_probability", 0) * 100 # Convert to percentage
human_prob = ensemble_data.get("human_probability", 0) * 100 # Convert to percentage
mixed_prob = ensemble_data.get("mixed_probability", 0) * 100 # Convert to percentage
confidence = ensemble_data.get("overall_confidence", 0) * 100 # Convert to percentage
uncertainty = ensemble_data.get("uncertainty_score", 0) * 100 # Convert to percentage
consensus = ensemble_data.get("consensus_level", 0) * 100 # Convert to percentage
final_verdict = ensemble_data.get("final_verdict", "Unknown")
total_time = performance_data.get("total_time", 0)
# Determine colors based on verdict
if ("Human".lower() in final_verdict.lower()):
verdict_color = SUCCESS_COLOR
elif ("AI".lower() in final_verdict.lower()):
verdict_color = DANGER_COLOR
elif ("Mixed".lower() in final_verdict.lower()):
verdict_color = WARNING_COLOR
else:
verdict_color = PRIMARY_COLOR
# PAGE 1: Analyzed File, Verdict, Reasoning, Key Indicators
# Header
header_style = ParagraphStyle('HeaderStyle',
parent = styles['Normal'],
fontName = 'Helvetica-Bold',
fontSize = 10,
textColor = GRAY_DARK,
alignment = TA_RIGHT,
)
elements.append(Paragraph("AI DETECTION ANALYTICS", header_style))
elements.append(HRFlowable(width = "100%",
thickness = 1,
color = PRIMARY_COLOR,
spaceAfter = 15,
)
)
# Title and main sections
elements.append(Paragraph("AI Text Detection Analysis Report", title_style))
elements.append(Paragraph(f"Generated on {datetime.now().strftime('%B %d, %Y at %I:%M %p')}", subtitle_style))
# Add original filename
elements.append(Paragraph(f"Analyzed File: {original_filename}", filename_style))
elements.append(Spacer(1, 0.1*inch))
# Add decorative line
elements.append(HRFlowable(width = "80%",
thickness = 2,
color = PRIMARY_COLOR,
spaceBefore = 10,
spaceAfter = 25,
hAlign = 'CENTER',
)
)
# Quick Stats Banner
stats_data = [['Text Source', 'AI', 'HUMAN', 'MIXED'],
['Probability', f"{ai_prob:.1f}%", f"{human_prob:.1f}%", f"{mixed_prob:.1f}%"]
]
stats_table = Table(stats_data, colWidths = [1.5*inch, 1*inch, 1*inch, 1*inch])
stats_table.setStyle(TableStyle([('BACKGROUND', (0, 0), (-1, 0), PRIMARY_COLOR),
('TEXTCOLOR', (0, 0), (-1, 0), colors.white),
('BACKGROUND', (1, 1), (1, 1), DANGER_COLOR),
('BACKGROUND', (2, 1), (2, 1), SUCCESS_COLOR),
('BACKGROUND', (3, 1), (3, 1), WARNING_COLOR),
('TEXTCOLOR', (1, 1), (-1, 1), colors.white),
('ALIGN', (0, 0), (-1, -1), 'CENTER'),
('FONTNAME', (0, 0), (-1, 0), 'Helvetica-Bold'),
('FONTSIZE', (0, 0), (-1, -1), 11),
('BOTTOMPADDING', (0, 0), (-1, -1), 8),
('TOPPADDING', (0, 0), (-1, -1), 8),
('GRID', (0, 0), (-1, -1), 0.5, colors.white),
('BOX', (0, 0), (-1, -1), 1, PRIMARY_COLOR),
])
)
elements.append(stats_table)
elements.append(Spacer(1, 0.3*inch))
# Main Verdict Section
elements.append(Paragraph("DETECTION VERDICT", section_style))
verdict_box_data = [[Paragraph(f"<font size=18 color='{verdict_color}'><b>{final_verdict.upper()}</b></font>", ParagraphStyle('VerdictText', alignment=TA_CENTER)),
Paragraph(f"<font size=12>Confidence: <b>{confidence:.1f}%</b></font><br/>"
f"<font size=10>Uncertainty: {uncertainty:.1f}% | Consensus: {consensus:.1f}%</font>",
ParagraphStyle('VerdictDetails', alignment=TA_CENTER))
]]
verdict_box = Table(verdict_box_data, colWidths = [2.5*inch, 3*inch])
verdict_box.setStyle(TableStyle([('BACKGROUND', (0, 0), (0, 0), GRAY_LIGHT),
('BACKGROUND', (1, 0), (1, 0), GRAY_LIGHT),
('BOX', (0, 0), (-1, -1), 1, verdict_color),
('ROUNDEDCORNERS', [10, 10, 10, 10]),
('ALIGN', (0, 0), (-1, -1), 'CENTER'),
('VALIGN', (0, 0), (-1, -1), 'MIDDLE'),
('BOTTOMPADDING', (0, 0), (-1, -1), 15),
('TOPPADDING', (0, 0), (-1, -1), 15),
])
)
elements.append(verdict_box)
elements.append(Spacer(1, 0.3*inch))
# DETECTION REASONING
elements.append(Paragraph("DETECTION REASONING", section_style))
# Process summary text and convert to bullet points
summary_text = reasoning.summary if hasattr(reasoning, 'summary') else "No reasoning summary available."
# Fix extra spaces first
summary_text = ' '.join(summary_text.split())
# Convert **bold** markers to HTML bold tags
summary_text = re.sub(r'\*\*(.*?)\*\*', r'<b>\1</b>', summary_text)
# Split into sentences and create bullet points
sentences = re.split(r'(?<=[.!?])\s+', summary_text)
# Create bullet points
for i, sentence in enumerate(sentences):
if sentence.strip():
# Add bullet point
elements.append(Paragraph(f"<font color='{PRIMARY_COLOR}'>•</font> {sentence.strip()}", bullet_style))
# Add extra spacing after each bullet point (except the last one)
if (i < len(sentences) - 1):
# Add spacing between bullet points
elements.append(Spacer(1, 0.08*inch))
# KEY INDICATORS
if ((hasattr(reasoning, 'key_indicators')) and reasoning.key_indicators and (len(reasoning.key_indicators) > 0)):
elements.append(Paragraph("KEY INDICATORS", key_indicators_style))
for indicator in reasoning.key_indicators:
if isinstance(indicator, str):
# Fix extra spaces
indicator = ' '.join(indicator.split())
# Convert **bold** markers to proper HTML bold tags
formatted_indicator = re.sub(r'\*\*(.*?)\*\*', r'<b>\1</b>', indicator)
# Fix underscores in metric names
formatted_indicator = formatted_indicator.replace('_', ' ')
elements.append(Paragraph(f"<font color='{SUCCESS_COLOR}'>•</font> {formatted_indicator}", body_style))
elements.append(Spacer(1, 0.05*inch))
elements.append(PageBreak())
# PAGE 2: Content Analysis & Metric Contributions
# CONTENT ANALYSIS
elements.append(Paragraph("CONTENT ANALYSIS", section_style))
domain = analysis_data.get("domain", "general").replace('_', ' ').upper()
# Convert to percentage
domain_confidence = analysis_data.get("domain_confidence", 0) * 100
text_length = analysis_data.get("text_length", 0)
sentence_count = analysis_data.get("sentence_count", 0)
# Create two-column layout for content analysis
content_data = [[Paragraph("<b>Content Domain</b>", bold_style), Paragraph(f"<font color='{INFO_COLOR}'><b>{domain}</b></font> ({domain_confidence:.1f}% confidence)", body_style)],
[Paragraph("<b>Text Statistics</b>", bold_style), Paragraph(f"{text_length:,} words | {sentence_count:,} sentences", body_style)],
[Paragraph("<b>Processing Time</b>", bold_style), Paragraph(f"{total_time:.2f} seconds", body_style)],
[Paragraph("<b>Analysis Method</b>", bold_style), Paragraph("Confidence-Weighted Ensemble Aggregation", body_style)],
]
content_table = Table(content_data, colWidths = [2*inch, 4.5*inch])
content_table.setStyle(TableStyle([('FONTNAME', (0, 0), (0, -1), 'Helvetica-Bold'),
('FONTNAME', (1, 0), (1, -1), 'Helvetica'),
('FONTSIZE', (0, 0), (-1, -1), 11),
('BOTTOMPADDING', (0, 0), (-1, -1), 10),
('TOPPADDING', (0, 0), (-1, -1), 10),
('GRID', (0, 0), (-1, -1), 0.25, GRAY_MEDIUM),
('BACKGROUND', (0, 0), (0, -1), GRAY_LIGHT),
])
)
elements.append(content_table)
elements.append(Spacer(1, 0.4*inch))
# METRIC CONTRIBUTIONS
elements.append(Paragraph("METRIC CONTRIBUTIONS", section_style))
metric_contributions = ensemble_data.get("metric_contributions", {})
if (metric_contributions and (len(metric_contributions) > 0)):
# Create clean table with updated headers
weight_data = [['METRIC NAME', 'ENSEMBLE WEIGHT (%)']]
for metric_name, contribution in metric_contributions.items():
weight = contribution.get("weight", 0) * 100
display_name = metric_name.replace('_', ' ').title()
weight_data.append([Paragraph(display_name, bold_style), Paragraph(f"{weight:.1f}%", body_style)])
# Setup Table Columns
weight_table = Table(weight_data, colWidths = [4*inch, 2.5*inch])
weight_table.setStyle(TableStyle([('BACKGROUND', (0, 0), (-1, 0), PRIMARY_COLOR),
('TEXTCOLOR', (0, 0), (-1, 0), colors.white),
('ALIGN', (0, 0), (-1, -1), 'LEFT'),
('ALIGN', (1, 0), (1, -1), 'RIGHT'),
('FONTNAME', (0, 0), (-1, 0), 'Helvetica-Bold'),
('FONTSIZE', (0, 0), (-1, -1), 11),
('BOTTOMPADDING', (0, 0), (-1, -1), 10),
('TOPPADDING', (0, 0), (-1, -1), 10),
('GRID', (0, 0), (-1, -1), 0.5, GRAY_MEDIUM),
('BACKGROUND', (1, 1), (1, -1), GRAY_LIGHT),
])
)
elements.append(weight_table)
# Add some filler content to reduce white space
elements.append(Spacer(1, 0.4*inch))
elements.append(HRFlowable(width = "100%", thickness = 1, color = PRIMARY_COLOR, spaceBefore = 10, spaceAfter = 10))
elements.append(Paragraph("<i>Report continues with detailed metric analysis on the following pages...</i>",
ParagraphStyle('ContinueStyle', parent = body_style, fontSize = 10, textColor = GRAY_DARK, alignment = TA_CENTER)))
elements.append(PageBreak())
# PAGE 3: STRUCTURAL & ENTROPY
elements.append(Paragraph("DETAILED METRIC ANALYSIS", section_style))
elements.append(Spacer(1, 0.2*inch))
# Filter for STRUCTURAL and ENTROPY only
page3_metrics = [m for m in detailed_metrics if m.name in ['structural', 'entropy']]
for metric in page3_metrics:
self._add_detailed_metric_section(elements = elements,
metric = metric,
small_bold_style = small_bold_style,
small_style = small_style,
bold_style = bold_style,
PRIMARY_COLOR = PRIMARY_COLOR,
SUCCESS_COLOR = SUCCESS_COLOR,
DANGER_COLOR = DANGER_COLOR,
WARNING_COLOR = WARNING_COLOR,
GRAY_LIGHT = GRAY_LIGHT,
)
elements.append(Spacer(1, 0.1*inch))
elements.append(HRFlowable(width = "100%", thickness = 0.5, color = GRAY_MEDIUM, spaceBefore = 5, spaceAfter = 15))
elements.append(PageBreak())
# PAGE 4: PERPLEXITY & SEMANTIC ANALYSIS
elements.append(Paragraph("DETAILED METRIC ANALYSIS", section_style))
elements.append(Spacer(1, 0.2*inch))
# Filter for PERPLEXITY and SEMANTIC_ANALYSIS only
page4_metrics = [m for m in detailed_metrics if m.name in ['perplexity', 'semantic_analysis']]
for metric in page4_metrics:
self._add_detailed_metric_section(elements = elements,
metric = metric,
small_bold_style = small_bold_style,
small_style = small_style,
bold_style = bold_style,
PRIMARY_COLOR = PRIMARY_COLOR,
SUCCESS_COLOR = SUCCESS_COLOR,
DANGER_COLOR = DANGER_COLOR,
WARNING_COLOR = WARNING_COLOR,
GRAY_LIGHT = GRAY_LIGHT,
)
elements.append(Spacer(1, 0.3*inch))
elements.append(HRFlowable(width = "100%", thickness = 0.5, color = GRAY_MEDIUM, spaceBefore = 5, spaceAfter = 15))
elements.append(PageBreak())
# PAGE 5: LINGUISTIC & MULTI PERTURBATION STABILITY
elements.append(Paragraph("DETAILED METRIC ANALYSIS", section_style))
elements.append(Spacer(1, 0.1*inch))
# Filter for LINGUISTIC and MULTI_PERTURBATION_STABILITY only
page5_metrics = [m for m in detailed_metrics if m.name in ['linguistic', 'multi_perturbation_stability']]
# Create a list to hold all content for Page 5
page5_elements = list()
for i, metric in enumerate(page5_metrics):
# Create temporary elements list for this metric
metric_elements = list()
# Add metric section to temporary list
self._add_detailed_metric_section(elements = metric_elements,
metric = metric,
small_bold_style = small_bold_style,
small_style = small_style,
bold_style = bold_style,
PRIMARY_COLOR = PRIMARY_COLOR,
SUCCESS_COLOR = SUCCESS_COLOR,
DANGER_COLOR = DANGER_COLOR,
WARNING_COLOR = WARNING_COLOR,
GRAY_LIGHT = GRAY_LIGHT,
)
# Add to page5_elements
page5_elements.extend(metric_elements)
# Add separator if not the last metric
if (i < len(page5_metrics) - 1):
page5_elements.append(Spacer(1, 0.05*inch)) # Minimal spacing
page5_elements.append(HRFlowable(width = "100%", thickness = 0.5, color = GRAY_MEDIUM, spaceBefore = 5, spaceAfter = 10))
# Add all page 5 elements to main elements
elements.extend(page5_elements)
elements.append(PageBreak())
# PAGE 6: Model Attribution & Recommendations
# AI MODEL ATTRIBUTION
if attribution_result:
elements.append(Paragraph("AI MODEL ATTRIBUTION", section_style))
elements.append(Spacer(1, 0.1*inch))
predicted_model = getattr(attribution_result.predicted_model, 'value', str(attribution_result.predicted_model))
predicted_model = predicted_model.replace("_", " ").title()
attribution_confidence = getattr(attribution_result, 'confidence', 0) * 100
domain_used = getattr(attribution_result.domain_used, 'value', 'Unknown').upper()
# Professional attribution table
attribution_data = [[Paragraph("<b>Predicted Model</b>", bold_style), Paragraph(f"<font color='{INFO_COLOR}'><b>{predicted_model}</b></font>", bold_style)],
[Paragraph("<b>Attribution Confidence</b>", bold_style), Paragraph(f"<b>{attribution_confidence:.1f}%</b>", bold_style)],
[Paragraph("<b>Domain Used</b>", bold_style), Paragraph(f"<b>{domain_used}</b>", bold_style)]
]
attribution_table = Table(attribution_data, colWidths = [2.5*inch, 4*inch])
attribution_table.setStyle(TableStyle([('BACKGROUND', (0, 0), (0, -1), GRAY_LIGHT),
('FONTNAME', (0, 0), (0, -1), 'Helvetica-Bold'),
('FONTSIZE', (0, 0), (-1, -1), 11),
('BOTTOMPADDING', (0, 0), (-1, -1), 8),
('TOPPADDING', (0, 0), (-1, -1), 8),
('GRID', (0, 0), (-1, -1), 0.5, GRAY_MEDIUM),
('VALIGN', (0, 0), (-1, -1), 'MIDDLE'),
])
)
elements.append(attribution_table)
elements.append(Spacer(1, 0.2*inch))
# MODEL PROBABILITY DISTRIBUTION
model_probs = getattr(attribution_result, 'model_probabilities', {})
if (model_probs and (len(model_probs) > 0)):
elements.append(Paragraph("MODEL PROBABILITY DISTRIBUTION", subsection_style))
elements.append(Spacer(1, 0.05*inch))
# Get top models
sorted_models = sorted(model_probs.items(), key = lambda x: x[1], reverse = True)[:10]
prob_data = [['LANGUAGE MODEL NAME', 'ATTRIBUTION PROBABILITY']]
for model_name, probability in sorted_models:
display_name = model_name.replace("_", " ").replace("-", " ").title()
prob_data.append([Paragraph(display_name, bold_style), Paragraph(f"{probability:.1%}", bold_style)])
# Table Columns Setup
prob_table = Table(prob_data, colWidths = [4*inch, 2.5*inch])
prob_table.setStyle(TableStyle([('BACKGROUND', (0, 0), (-1, 0), INFO_COLOR),
('TEXTCOLOR', (0, 0), (-1, 0), colors.white),
('ALIGN', (0, 0), (-1, -1), 'LEFT'),
('ALIGN', (1, 0), (1, -1), 'RIGHT'),
('FONTNAME', (0, 0), (-1, 0), 'Helvetica-Bold'),
('FONTSIZE', (0, 0), (-1, -1), 11),
('BOTTOMPADDING', (0, 0), (-1, -1), 6),
('TOPPADDING', (0, 0), (-1, -1), 6),
('GRID', (0, 0), (-1, -1), 0.5, GRAY_MEDIUM),
('BACKGROUND', (1, 1), (1, -1), GRAY_LIGHT),
])
)
elements.append(prob_table)
elements.append(Spacer(1, 0.3*inch))
# RECOMMENDATIONS
if ((hasattr(reasoning, 'recommendations')) and reasoning.recommendations):
elements.append(Paragraph("RECOMMENDATIONS", section_style))
elements.append(Spacer(1, 0.1*inch))
for i, recommendation in enumerate(reasoning.recommendations):
# Alternate colors for visual interest
if (i % 3 == 0):
rec_color = SUCCESS_COLOR
elif (i % 3 == 1):
rec_color = INFO_COLOR
else:
rec_color = WARNING_COLOR
# Clean up recommendation text - fix spaces and bold markers
clean_rec = ' '.join(recommendation.split())
clean_rec = re.sub(r'\*\*(.*?)\*\*', r'<b>\1</b>', clean_rec)
clean_rec = clean_rec.replace('_', ' ')
rec_box_data = [[Paragraph(f"<font color='{rec_color}'>✓</font> {clean_rec}", body_style)]]
rec_box = Table(rec_box_data, colWidths = [6.5*inch])
rec_box.setStyle(TableStyle([('BACKGROUND', (0, 0), (-1, -1), GRAY_LIGHT),
('BOX', (0, 0), (-1, -1), 1, rec_color),
('PADDING', (0, 0), (-1, -1), 10),
('LEFTPADDING', (0, 0), (-1, -1), 8),
('BOTTOMMARGIN', (0, 0), (-1, -1), 6),
])
)
elements.append(rec_box)
elements.append(Spacer(1, 0.2*inch))
# Footer with watermark
elements.append(Spacer(1, 0.2*inch))
elements.append(HRFlowable(width = "100%", thickness = 0.5, color = GRAY_MEDIUM, spaceAfter = 8))
# Extract report ID from filename
report_id = filename.replace('.pdf', '')
footer_text = (f"Generated by AI Text Detector v1.0 | "
f"Processing Time: {total_time:.2f}s | "
f"Report ID: {report_id}")
elements.append(Paragraph(footer_text, footer_style))
elements.append(Paragraph("Confidential Analysis Report • © 2025 AI Detection Analytics",
ParagraphStyle('Copyright', parent = footer_style, fontSize = 8, textColor = GRAY_MEDIUM)))
# Build PDF
doc.build(elements)
logger.info(f"PDF report saved: {output_path}")
return output_path
def _add_detailed_metric_section(self, elements, metric, small_bold_style, small_style, bold_style, PRIMARY_COLOR, SUCCESS_COLOR, DANGER_COLOR, WARNING_COLOR, GRAY_LIGHT):
"""
Add a detailed metric section to the PDF
"""
# Import needed components
from reportlab.platypus import Paragraph, Table, Spacer
from reportlab.platypus import TableStyle
from reportlab.lib import colors
from reportlab.lib.units import inch
from reportlab.lib.styles import ParagraphStyle
from reportlab.lib.enums import TA_LEFT
# Determine metric color based on verdict
if (metric.verdict == "HUMAN"):
metric_color = SUCCESS_COLOR
prob_color = SUCCESS_COLOR
elif (metric.verdict == "AI"):
metric_color = DANGER_COLOR
prob_color = DANGER_COLOR
else:
metric_color = WARNING_COLOR
prob_color = WARNING_COLOR
# Create professional metric header
metric_display_name = metric.name.replace('_', ' ').upper()
# Metric title and description
subsection_style = ParagraphStyle('SubsectionStyle',
parent = ParagraphStyle('Normal'),
fontName = 'Helvetica-Bold',
fontSize = 14,
textColor = PRIMARY_COLOR,
spaceAfter = 8,
spaceBefore = 16,
alignment=TA_LEFT,
)
elements.append(Paragraph(f"<b>{metric_display_name}</b>", subsection_style))
elements.append(Paragraph(f"<i>{metric.description}</i>", small_style))
elements.append(Spacer(1, 0.1*inch))
# Key metrics in a clean table
key_metrics_data = [[Paragraph("<b>Verdict</b>", bold_style), Paragraph(f"<font color='{metric_color}'><b>{metric.verdict}</b></font>", bold_style), Paragraph("<b>Weight</b>", bold_style), Paragraph(f"<b>{metric.weight:.1f}%</b>", bold_style)],
[Paragraph("<b>AI Probability</b>", bold_style), Paragraph(f"<font color='{prob_color}'><b>{metric.ai_probability:.1f}%</b></font>", bold_style), Paragraph("<b>Confidence</b>", bold_style), Paragraph(f"<b>{metric.confidence:.1f}%</b>", bold_style)]
]
key_metrics_table = Table(key_metrics_data, colWidths = [1.5*inch, 1.5*inch, 1.5*inch, 1.5*inch])
key_metrics_table.setStyle(TableStyle([('BACKGROUND', (0, 0), (-1, -1), GRAY_LIGHT),
('GRID', (0, 0), (-1, -1), 0.5, colors.white),
('VALIGN', (0, 0), (-1, -1), 'MIDDLE'),
('BOTTOMPADDING', (0, 0), (-1, -1), 8),
('TOPPADDING', (0, 0), (-1, -1), 8),
('ALIGN', (0, 0), (-1, -1), 'CENTER'),
])
)
elements.append(key_metrics_table)
elements.append(Spacer(1, 0.2*inch))
# Detailed metrics in a compact table
if metric.detailed_metrics and len(metric.detailed_metrics) > 0:
# Create table with all metrics
detailed_data = []
# Sort metrics alphabetically
sorted_items = sorted(metric.detailed_metrics.items())
# Group into rows with 3 metrics per row
for i in range(0, len(sorted_items), 3):
row = []
# Add up to 3 metrics per row
for j in range(3):
if i + j < len(sorted_items):
key, value = sorted_items[i + j]
# Format key name properly
display_key = key.replace('_', ' ').title()
formatted_value = self._format_metric_value(key, value)
row.append(Paragraph(f"<font size=9><b>{display_key}:</b></font>", small_bold_style))
row.append(Paragraph(f"<font size=9>{formatted_value}</font>", small_style))
else:
row.append("")
row.append("")
detailed_data.append(row)
if detailed_data:
# Calculate column widths dynamically
col_width = 6.5 * inch / 6 # 6 columns total
col_widths = [col_width] * 6
detailed_table = Table(detailed_data, colWidths = col_widths)
detailed_table.setStyle(TableStyle([('FONTSIZE', (0, 0), (-1, -1), 8),
('BOTTOMPADDING', (0, 0), (-1, -1), 3),
('TOPPADDING', (0, 0), (-1, -1), 3),
('GRID', (0, 0), (-1, -1), 0.2, colors.grey),
('VALIGN', (0, 0), (-1, -1), 'MIDDLE'),
('ALIGN', (1, 0), (1, -1), 'RIGHT'),
('ALIGN', (3, 0), (3, -1), 'RIGHT'),
('ALIGN', (5, 0), (5, -1), 'RIGHT'),
])
)
elements.append(detailed_table)
def _format_metric_value(self, key: str, value: Any) -> str:
"""
Format metric value based on its type
"""
if not isinstance(value, (int, float)):
return str(value)
key_lower = key.lower()
if ('perplexity' in key_lower):
if (value > 1000):
return f"{value:,.0f}"
else:
return f"{value:.2f}"
elif (('probability' in key_lower) or ('confidence' in key_lower)):
return f"{value:.1f}%"
elif ('entropy' in key_lower):
return f"{value:.2f}"
elif (('ratio' in key_lower) or ('score' in key_lower)):
if (0 <= value <= 1):
return f"{value:.3f}"
else:
return f"{value:.2f}"
elif (key_lower in ['num_sentences', 'num_words', 'vocabulary_size']):
return f"{int(value):,}"
elif (('length' in key_lower) or ('size' in key_lower)):
return f"{value:.2f}"
else:
return f"{value:.3f}"
# Export
__all__ = ["ReportGenerator",
"DetailedMetric",
] |